The Internet is a system of geographically distributed computer networks interconnected by computers executing networking protocols that allow users to interact and share information over the networks. Because of such wide-spread information sharing, the Internet has generally evolved into an "open" system for which developers can design software for performing specialized operations, or services, essentially without restriction. These services are typically implemented in accordance with a client/server architecture, wherein the clients, e.g., personal computers or workstations, are responsible for interacting with the users and the servers are computers configured to perform the services as directed by the clients.
Not surprisingly, each of the services available over the Internet is generally defined by its own networking protocol. A protocol is a set of rules governing the format and meaning of messages or "packets" exchanged over the networks. By implementing services in accordance with the protocols, computers cooperate to perform various operations, or similar operations in various ways, for users wishing to "interact" with the networks. The services typically range from browsing or searching for information having a particular data format using a particular protocol to actually acquiring information of a different format in accordance with a different protocol.
For example, the file transfer protocol (FTP) service facilitates the transfer and sharing of files across the Internet. The Telnet service allows users to log onto computers coupled to the networks, while the netnews protocol provides a bulletin-board service to its subscribers. Furthermore, the various data formats of the information available on the Internet include JPEG images, MPEG movies and .mu.-law sound files.
Coincident with the design of these services has been the development of applications for implementing the services on the client/server architecture. Accordingly, applications are available for users to obtain files from computers connected to the Internet using the FTP protocol. Similarly, individual applications allow users to log into remote computers (as though they were logging in from terminals attached to those computers) using the Telnet protocol and, further, to view JPEG images and MPEG movies. As a result, there exists a proliferation of applications directed to user activity on the Internet.
A problem with this vast collection of application-specific protocols is that these applications are generally unorganized, thus requiring users to plod through them in order to satisfyingly, and profitably, utilize the Internet. Such lack of uniformity is time consuming and disorienting to users that want to access particular types of information but are forced to use unfamiliar applications. Because of the enormous amount of different types of information available on the Internet and the variety of applications needed to access those information types, the experience of using the Internet may be burdensome to these users.
An alternative to the assortment of open applications for accessing information on the Internet is a "closed" application system, such as Prodigy, CompuServe or America Online. Each of these systems provide a full range of well-organized services to their subscribers; however, they also impose restrictions on the services developers can offer for their systems. Such constraint of "new" service development may be an unreasonable alternative for many users.
Two fashionable services for accessing information over the Internet are Gopher and the World-Wide Web ("Web"). Gopher consists of a series of Internet servers that provide a "list-oriented" interface to information available on the networks; the information is displayed as menu items in a hierarchical manner. Included in the hierarchy of menus are documents, which can be displayed or saved, and searchable indexes, which allow users to type keywords and perform searches.
Some of the menu items displayed by Gopher are links to information available on other servers located on the networks. In this case, the user is presented with a list of available information documents that can be opened. The opened documents may display additional lists or they may contain various data-types, such as pictures or text; occasionally, the opened documents may "transport" the user to another computer on the Internet.
The other popular information service on the Internet is the Web. Instead of providing a user with a hierarchical list-oriented view of information, the Web provides the user with a "linked-hypertext" view. Metaphorically, the Web perceives the Internet as a vast book of pages, each of which may contain pictures, text, sound, movies or various other types of data in the form of documents. Web documents are written in HyperText Markup Language (HTML) and Web servers transfer HTML documents to each other through the HyperText Transfer Protocol (HTTP).
The Web service is essentially a means for naming sources of information on the Internet. Armed with such a general naming convention that spans the entire network system, developers are able to build information servers that potentially any user can access. Accordingly, Gopher servers, HTTP servers, FTP servers, and E-mail servers have been developed for the Web. Moreover, the naming convention enables users to identify resources (such as directories and documents) on any of these servers connected to the Internet and allow access to those resources.
As an example, a user "traverses" the Web by following hot items of a page displayed on a graphical Web browser. These hot items are hypertext links whose presence are indicated on the page by visual cues, e.g., underlined words, icons or buttons. When a user follows a link (usually by clicking on the cue with a mouse), the browser displays the target pointed to by the link which, in some cases, may be another HTML document.
The Gopher and Web information services represent entirely different approaches to interacting with information on the Internet. One follows a list-approach to information that "looks" like a telephone directory service, while the other assumes a page-approach analogous to a tabloid newspaper. However, both of these approaches include applications for enabling users to browse information available on Internet servers. For example, TurboGopher is a text-based browser application for client computers operating in accordance with the Gopher protocol. Additionally, each of these applications has a unique way of viewing and accessing the information on server computers.
Typically, a computer includes an operating system and application software which, collectively, control the operations of the computer. The applications are preferably task-specific and independent, e.g., a word processor application edits text, a drawing application edits drawings and a database application interacts with information stored on a database storage unit. Although a user can move data from one application to the other, such as by copying a drawing into a word processing file, the independent applications must be invoked to thereafter manipulate that data. This is primarily because the applications are designed as monolithic, autonomous modules that generally do not interact efficiently.
Netscape Navigator.TM. ("Netscape") is an example of a monolithic Web browser application that is configured to interact with many of the previously-described protocols, including HTTP, Gopher and FTP. However when instructed to invoke an application that uses one of these protocols, Netscape "translates" the protocol to hypertext. Such awkward translation places the user farther away from the protocol designed to run the application and, in some cases, actually thwarts the user's Internet experience. For example, a discussion system requiring an interactive exchange between participants may be bogged down by hypertext translations.
Each of these information services provides a mechanism for keeping track of places explored while navigating the computer networks. For example, Netscape has a log mechanism which typically manifests as a pop-up menu that shows recent places a user has explored. These mechanisms, however, are not generally configured to interact with applications other than their own, primarily due to the monolithic nature of the Netscape and TurboGopher applications. Thus, it is difficult to replace the logs with similar mechanisms provided by other developers; furthermore, the monolithic nature of the applications generally precludes extension of the functions and behaviors of these mechanisms other than by those developers associated with the respective services.
The Gopher and Web services may further require additional applications to perform specific functions, such as playing sound or viewing movies, with respect to the data types contained in the documents. For example, Netscape employs helper applications for executing application programs having data formats it does not "understand". Execution of these functions on a computer requires interruption of processing and context switching (i.e., saving of state) prior to invoking the appropriate application. Thus, if a user operating within the Netscape application "opens" a MPEG movie, that browsing application must be saved (e.g., to disk) prior to opening an appropriate MPEG application, e.g., Sparkle, to view the image. Such an arrangement is inefficient and rather disruptive to processing operations of the computer.
Generally, the application program presents information to a user through a window of a graphical user interface by drawing images, graphics or text within the window region. The user, in turn, communicates with the application by "pointing" at graphical objects in the window with a pointer that is controlled by a hand-operated pointing device, such as a mouse, or by pressing keys of a keyboard.
The graphical objects typically included with each window region are sizing boxes, buttons and scroll bars. These objects represent user interface elements that the user can point at with the pointer (or a cursor) to select or manipulate. For example, the user may manipulate these elements to move the windows around on the display screen, and change their sizes and appearances so as to arrange the window in a convenient manner. When the elements are selected or manipulated, the underlying application program is informed, via the window environment, that control has been appropriated by the user.
A menu bar is a further example of a user interface element that provides a list of menus available to a user. Each menu, in turn, provides a list of command options that can be selected merely by pointing to them with the mouse-controlled pointer. That is, the commands may be issued by actuating the mouse to move the pointer onto or near the command selection, and pressing and quickly releasing, i.e., "clicking" a button on the mouse.
Command options in the menus typically act on one or more objects. If an application requires more information to carry out a command, a dialog box may be employed. A dialog box is a special type of window that elicits information from the user, such as requiring the user to check items or fill in the blanks, typically through the use of a graphical object that the user can manipulate with the mouse. Such an object is called a control and, by manipulating the control, the user can take an immediate action or change a setting to modify a future action. A modal dialog box requires that the user respond to the elicitation prior to performing other actions.
In contrast to this typical application-based computing environment, a software component architecture provides a modular document-based computing arrangement using tools such as viewing editors. The key to document-based computing is the compound document, i.e., a document composed of many different types of data sharing the same file. The types of data contained in a compound document may range from text, tables and graphics to video and sound. Several editors, each designed to handle a particular data type or format, can work on the contents of the document at the same time, unlike the application-based computing environment.
Since many editors may work together on the same document, the compound document is apportioned into individual modules of content for manipulation by the editors. The compound-nature of the document is realized by embedding these modules within each other to create a document having a mixture of data types. The software component architecture provides the foundation for assembling documents of differing contents and the present invention is directed to a system for extending this capability to network-oriented services.
Therefore, it is among the objects of the present invention to simplify a user's experience on computer networks without sacrificing the flexibility afforded the user by employing existing protocols and data types available on those networks.
Another object of the invention is to provide a document-based computing system that enables users to replace and extend modules configured to provide services directed to information available on computer networks.
Still another object of the invention is to provide a framework that allows third-party developers to extend a layered network component system by building new components that seamlessly interact with the system components.